518 A TREATISE ON METAMORPHISM. 



of the 'constituent might have been abstracted, and thus the amount of 

 residual material left be but 10 per cent or 1 per cent of the original amount. 

 Doubtless even the latter shrinkage sometimes occurs in the case of the 

 limestones. It is therefore manifest that all that the foregoing tables show 

 is that the losses calculated for the elements are at least as great as given, 

 and in many cases they are in all probability much greater. Probably 

 more accurate calculations than have been made are possible by taking as 

 the constant constituents the very difficultly decomposable accessory min- 

 erals. For instance, where zircon is present and is uniformly distributed 

 through the rock the assumption that this mineral is constant would 

 undoubtedly lead to closer approximation to truth than has been obtained on 

 the hypothesis that the iron or aluminum, or both, are constant. 



But in order fully to solve the problem of the amount of losses of the 

 various elements, in addition to the facts given in the tables, we need to 

 know the original mass of the material as compared with the resultant mass. 

 If we could combine these data we could give a quantitative estimate of the 

 loss of each of the elements due to solution in the belt of weathering. But 

 as long as one of the two necessary factors of the problem is not ascertained, 

 we know only that the losses of material are much greater than any method 

 of calculation yet made would lead us to believe. 



ORDER OF DECOMPOSITION OF THE MINERALS, AND THE END PRODUCTS. 



From the relative amounts of losses of the various elements," as given on 

 page 515, one may infer the relative rate of decomposition of the minerals. 

 The alkalies sodium and potassium commonly occur together; the same is 

 true for the alkaline earths calcium and magnesium; but neither of these 

 rules is invariable. One would expect the heavily alkaline minerals to be 

 decomposed most rapidly; next in order the alkaline earth minerals; next 

 in order the ferromagnesian minerals; and lastly and least rapidly the 

 heavily aluminous minerals. Of course this statement is only of the most 

 general kind, since most of the abundant minerals may have as bases more 

 than one of these groups. But in proportion as the alkali end of the series 

 dominates, the minerals are soluble; in proportion as the aluminous end of 

 the series is dominant, they are insoluble. 



We thus have the explanation of the order of decomposability of the 

 silicates. Among the common anhydrous minerals, liephelite-leucite min- 

 erals are the most readily decomposed Second in order of decomposability 



